| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 4.0, |
| "eval_steps": 500, |
| "global_step": 200, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 0.699141651391983, |
| "epoch": 0.02, |
| "grad_norm": 1.7043750286102295, |
| "learning_rate": 0.0, |
| "loss": 1.5967991352081299, |
| "mean_token_accuracy": 0.665926069021225, |
| "num_tokens": 74699.0, |
| "step": 1 |
| }, |
| { |
| "entropy": 0.6601996421813965, |
| "epoch": 0.04, |
| "grad_norm": 1.6597099304199219, |
| "learning_rate": 1e-05, |
| "loss": 1.5338245630264282, |
| "mean_token_accuracy": 0.6820768415927887, |
| "num_tokens": 138614.0, |
| "step": 2 |
| }, |
| { |
| "entropy": 0.7289904654026031, |
| "epoch": 0.06, |
| "grad_norm": 1.796127438545227, |
| "learning_rate": 2e-05, |
| "loss": 1.6285251379013062, |
| "mean_token_accuracy": 0.6539906561374664, |
| "num_tokens": 209509.0, |
| "step": 3 |
| }, |
| { |
| "entropy": 0.7412302494049072, |
| "epoch": 0.08, |
| "grad_norm": 1.4846091270446777, |
| "learning_rate": 3e-05, |
| "loss": 1.5454721450805664, |
| "mean_token_accuracy": 0.6691758334636688, |
| "num_tokens": 283751.0, |
| "step": 4 |
| }, |
| { |
| "entropy": 0.7322555780410767, |
| "epoch": 0.1, |
| "grad_norm": 1.4136911630630493, |
| "learning_rate": 4e-05, |
| "loss": 1.5077476501464844, |
| "mean_token_accuracy": 0.6744928658008575, |
| "num_tokens": 354066.0, |
| "step": 5 |
| }, |
| { |
| "entropy": 0.7739330530166626, |
| "epoch": 0.12, |
| "grad_norm": 0.9657893180847168, |
| "learning_rate": 5e-05, |
| "loss": 1.43147611618042, |
| "mean_token_accuracy": 0.6826294660568237, |
| "num_tokens": 422170.0, |
| "step": 6 |
| }, |
| { |
| "entropy": 0.8858513236045837, |
| "epoch": 0.14, |
| "grad_norm": 0.7331474423408508, |
| "learning_rate": 6e-05, |
| "loss": 1.450822353363037, |
| "mean_token_accuracy": 0.6648280322551727, |
| "num_tokens": 494040.0, |
| "step": 7 |
| }, |
| { |
| "entropy": 0.9634815156459808, |
| "epoch": 0.16, |
| "grad_norm": 0.5526663661003113, |
| "learning_rate": 7e-05, |
| "loss": 1.3808112144470215, |
| "mean_token_accuracy": 0.6719212830066681, |
| "num_tokens": 566616.0, |
| "step": 8 |
| }, |
| { |
| "entropy": 1.0627405643463135, |
| "epoch": 0.18, |
| "grad_norm": 0.48948273062705994, |
| "learning_rate": 8e-05, |
| "loss": 1.3656816482543945, |
| "mean_token_accuracy": 0.663449615240097, |
| "num_tokens": 637093.0, |
| "step": 9 |
| }, |
| { |
| "entropy": 1.1328747868537903, |
| "epoch": 0.2, |
| "grad_norm": 0.37054377794265747, |
| "learning_rate": 9e-05, |
| "loss": 1.2878801822662354, |
| "mean_token_accuracy": 0.6749334335327148, |
| "num_tokens": 701945.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 1.2100302577018738, |
| "epoch": 0.22, |
| "grad_norm": 0.3048469126224518, |
| "learning_rate": 0.0001, |
| "loss": 1.236545443534851, |
| "mean_token_accuracy": 0.6853303909301758, |
| "num_tokens": 774533.0, |
| "step": 11 |
| }, |
| { |
| "entropy": 1.2940571904182434, |
| "epoch": 0.24, |
| "grad_norm": 0.3031807243824005, |
| "learning_rate": 0.0001, |
| "loss": 1.211038589477539, |
| "mean_token_accuracy": 0.6858378350734711, |
| "num_tokens": 847636.0, |
| "step": 12 |
| }, |
| { |
| "entropy": 1.299929440021515, |
| "epoch": 0.26, |
| "grad_norm": 0.2886947691440582, |
| "learning_rate": 0.0001, |
| "loss": 1.197202205657959, |
| "mean_token_accuracy": 0.6924754083156586, |
| "num_tokens": 923420.0, |
| "step": 13 |
| }, |
| { |
| "entropy": 1.252187728881836, |
| "epoch": 0.28, |
| "grad_norm": 0.25853437185287476, |
| "learning_rate": 0.0001, |
| "loss": 1.1612238883972168, |
| "mean_token_accuracy": 0.6976969540119171, |
| "num_tokens": 992109.0, |
| "step": 14 |
| }, |
| { |
| "entropy": 1.220507800579071, |
| "epoch": 0.3, |
| "grad_norm": 0.24553938210010529, |
| "learning_rate": 0.0001, |
| "loss": 1.1456246376037598, |
| "mean_token_accuracy": 0.6981014311313629, |
| "num_tokens": 1065114.0, |
| "step": 15 |
| }, |
| { |
| "entropy": 1.198462724685669, |
| "epoch": 0.32, |
| "grad_norm": 0.24306544661521912, |
| "learning_rate": 0.0001, |
| "loss": 1.1577763557434082, |
| "mean_token_accuracy": 0.6941779851913452, |
| "num_tokens": 1136722.0, |
| "step": 16 |
| }, |
| { |
| "entropy": 1.1386296153068542, |
| "epoch": 0.34, |
| "grad_norm": 0.22727081179618835, |
| "learning_rate": 0.0001, |
| "loss": 1.0986605882644653, |
| "mean_token_accuracy": 0.7019487023353577, |
| "num_tokens": 1209030.0, |
| "step": 17 |
| }, |
| { |
| "entropy": 1.117907702922821, |
| "epoch": 0.36, |
| "grad_norm": 0.21270814538002014, |
| "learning_rate": 0.0001, |
| "loss": 1.110465407371521, |
| "mean_token_accuracy": 0.701825737953186, |
| "num_tokens": 1276624.0, |
| "step": 18 |
| }, |
| { |
| "entropy": 1.1316577792167664, |
| "epoch": 0.38, |
| "grad_norm": 0.1995684653520584, |
| "learning_rate": 0.0001, |
| "loss": 1.1611299514770508, |
| "mean_token_accuracy": 0.6892036497592926, |
| "num_tokens": 1348478.0, |
| "step": 19 |
| }, |
| { |
| "entropy": 1.0270569920539856, |
| "epoch": 0.4, |
| "grad_norm": 0.18478907644748688, |
| "learning_rate": 0.0001, |
| "loss": 1.04264235496521, |
| "mean_token_accuracy": 0.7172180116176605, |
| "num_tokens": 1421488.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 1.01737579703331, |
| "epoch": 0.42, |
| "grad_norm": 0.19284074008464813, |
| "learning_rate": 0.0001, |
| "loss": 1.0274319648742676, |
| "mean_token_accuracy": 0.718668669462204, |
| "num_tokens": 1488403.0, |
| "step": 21 |
| }, |
| { |
| "entropy": 1.0264459252357483, |
| "epoch": 0.44, |
| "grad_norm": 0.1576310396194458, |
| "learning_rate": 0.0001, |
| "loss": 1.0518046617507935, |
| "mean_token_accuracy": 0.7142763733863831, |
| "num_tokens": 1564657.0, |
| "step": 22 |
| }, |
| { |
| "entropy": 1.059712529182434, |
| "epoch": 0.46, |
| "grad_norm": 0.15488146245479584, |
| "learning_rate": 0.0001, |
| "loss": 1.082923412322998, |
| "mean_token_accuracy": 0.7053070664405823, |
| "num_tokens": 1633972.0, |
| "step": 23 |
| }, |
| { |
| "entropy": 1.0438059866428375, |
| "epoch": 0.48, |
| "grad_norm": 0.15164747834205627, |
| "learning_rate": 0.0001, |
| "loss": 1.0578590631484985, |
| "mean_token_accuracy": 0.7133743166923523, |
| "num_tokens": 1700893.0, |
| "step": 24 |
| }, |
| { |
| "entropy": 0.9979909062385559, |
| "epoch": 0.5, |
| "grad_norm": 0.14712916314601898, |
| "learning_rate": 0.0001, |
| "loss": 1.0294842720031738, |
| "mean_token_accuracy": 0.7216654419898987, |
| "num_tokens": 1767721.0, |
| "step": 25 |
| }, |
| { |
| "entropy": 1.0120922327041626, |
| "epoch": 0.52, |
| "grad_norm": 0.1360112428665161, |
| "learning_rate": 0.0001, |
| "loss": 1.0354883670806885, |
| "mean_token_accuracy": 0.7223899960517883, |
| "num_tokens": 1835156.0, |
| "step": 26 |
| }, |
| { |
| "entropy": 0.9896539449691772, |
| "epoch": 0.54, |
| "grad_norm": 0.12824475765228271, |
| "learning_rate": 0.0001, |
| "loss": 1.0203971862792969, |
| "mean_token_accuracy": 0.7212119996547699, |
| "num_tokens": 1910430.0, |
| "step": 27 |
| }, |
| { |
| "entropy": 0.9603240489959717, |
| "epoch": 0.56, |
| "grad_norm": 0.1179676204919815, |
| "learning_rate": 0.0001, |
| "loss": 0.9597353339195251, |
| "mean_token_accuracy": 0.7346547842025757, |
| "num_tokens": 1985373.0, |
| "step": 28 |
| }, |
| { |
| "entropy": 1.0493255257606506, |
| "epoch": 0.58, |
| "grad_norm": 0.12462859600782394, |
| "learning_rate": 0.0001, |
| "loss": 1.0499563217163086, |
| "mean_token_accuracy": 0.7120701968669891, |
| "num_tokens": 2054409.0, |
| "step": 29 |
| }, |
| { |
| "entropy": 1.0497854351997375, |
| "epoch": 0.6, |
| "grad_norm": 0.11791805922985077, |
| "learning_rate": 0.0001, |
| "loss": 1.0524396896362305, |
| "mean_token_accuracy": 0.7136416733264923, |
| "num_tokens": 2122893.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 1.0638166069984436, |
| "epoch": 0.62, |
| "grad_norm": 0.12578116357326508, |
| "learning_rate": 0.0001, |
| "loss": 1.067002534866333, |
| "mean_token_accuracy": 0.7095249891281128, |
| "num_tokens": 2194755.0, |
| "step": 31 |
| }, |
| { |
| "entropy": 1.099778175354004, |
| "epoch": 0.64, |
| "grad_norm": 0.12349284440279007, |
| "learning_rate": 0.0001, |
| "loss": 1.0939478874206543, |
| "mean_token_accuracy": 0.7056429386138916, |
| "num_tokens": 2262214.0, |
| "step": 32 |
| }, |
| { |
| "entropy": 1.0646151900291443, |
| "epoch": 0.66, |
| "grad_norm": 0.12846575677394867, |
| "learning_rate": 0.0001, |
| "loss": 1.0653318166732788, |
| "mean_token_accuracy": 0.7122560441493988, |
| "num_tokens": 2331419.0, |
| "step": 33 |
| }, |
| { |
| "entropy": 1.0158570408821106, |
| "epoch": 0.68, |
| "grad_norm": 0.11879603564739227, |
| "learning_rate": 0.0001, |
| "loss": 1.0267729759216309, |
| "mean_token_accuracy": 0.7170567810535431, |
| "num_tokens": 2403814.0, |
| "step": 34 |
| }, |
| { |
| "entropy": 1.0103608667850494, |
| "epoch": 0.7, |
| "grad_norm": 0.12001688033342361, |
| "learning_rate": 0.0001, |
| "loss": 1.0078749656677246, |
| "mean_token_accuracy": 0.7219332754611969, |
| "num_tokens": 2474066.0, |
| "step": 35 |
| }, |
| { |
| "entropy": 1.0085046291351318, |
| "epoch": 0.72, |
| "grad_norm": 0.11229556053876877, |
| "learning_rate": 0.0001, |
| "loss": 1.015348196029663, |
| "mean_token_accuracy": 0.7219782471656799, |
| "num_tokens": 2547283.0, |
| "step": 36 |
| }, |
| { |
| "entropy": 1.0116952955722809, |
| "epoch": 0.74, |
| "grad_norm": 0.12163597345352173, |
| "learning_rate": 0.0001, |
| "loss": 1.0121233463287354, |
| "mean_token_accuracy": 0.7187064588069916, |
| "num_tokens": 2613331.0, |
| "step": 37 |
| }, |
| { |
| "entropy": 0.9940676093101501, |
| "epoch": 0.76, |
| "grad_norm": 0.11075941473245621, |
| "learning_rate": 0.0001, |
| "loss": 0.992800235748291, |
| "mean_token_accuracy": 0.7255141139030457, |
| "num_tokens": 2683528.0, |
| "step": 38 |
| }, |
| { |
| "entropy": 1.0126079320907593, |
| "epoch": 0.78, |
| "grad_norm": 0.11469519883394241, |
| "learning_rate": 0.0001, |
| "loss": 0.9988532066345215, |
| "mean_token_accuracy": 0.7211731672286987, |
| "num_tokens": 2750069.0, |
| "step": 39 |
| }, |
| { |
| "entropy": 0.9906446635723114, |
| "epoch": 0.8, |
| "grad_norm": 0.11193952709436417, |
| "learning_rate": 0.0001, |
| "loss": 0.9900511503219604, |
| "mean_token_accuracy": 0.722189337015152, |
| "num_tokens": 2821887.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.9898229837417603, |
| "epoch": 0.82, |
| "grad_norm": 0.11920945346355438, |
| "learning_rate": 0.0001, |
| "loss": 0.9936800599098206, |
| "mean_token_accuracy": 0.7267130017280579, |
| "num_tokens": 2885811.0, |
| "step": 41 |
| }, |
| { |
| "entropy": 0.9716921448707581, |
| "epoch": 0.84, |
| "grad_norm": 0.11293631792068481, |
| "learning_rate": 0.0001, |
| "loss": 0.9800352454185486, |
| "mean_token_accuracy": 0.7265651822090149, |
| "num_tokens": 2955201.0, |
| "step": 42 |
| }, |
| { |
| "entropy": 0.9296073317527771, |
| "epoch": 0.86, |
| "grad_norm": 0.11454292386770248, |
| "learning_rate": 0.0001, |
| "loss": 0.9429284334182739, |
| "mean_token_accuracy": 0.734944611787796, |
| "num_tokens": 3019665.0, |
| "step": 43 |
| }, |
| { |
| "entropy": 0.9590237438678741, |
| "epoch": 0.88, |
| "grad_norm": 0.1102190762758255, |
| "learning_rate": 0.0001, |
| "loss": 0.9501940011978149, |
| "mean_token_accuracy": 0.7321249544620514, |
| "num_tokens": 3093335.0, |
| "step": 44 |
| }, |
| { |
| "entropy": 0.9053312838077545, |
| "epoch": 0.9, |
| "grad_norm": 0.11164271831512451, |
| "learning_rate": 0.0001, |
| "loss": 0.9103053212165833, |
| "mean_token_accuracy": 0.7460834383964539, |
| "num_tokens": 3159757.0, |
| "step": 45 |
| }, |
| { |
| "entropy": 0.9606629908084869, |
| "epoch": 0.92, |
| "grad_norm": 0.11079204827547073, |
| "learning_rate": 0.0001, |
| "loss": 0.9644097089767456, |
| "mean_token_accuracy": 0.7342032492160797, |
| "num_tokens": 3230892.0, |
| "step": 46 |
| }, |
| { |
| "entropy": 0.943544864654541, |
| "epoch": 0.94, |
| "grad_norm": 0.11624015867710114, |
| "learning_rate": 0.0001, |
| "loss": 0.9410842061042786, |
| "mean_token_accuracy": 0.7351740300655365, |
| "num_tokens": 3299869.0, |
| "step": 47 |
| }, |
| { |
| "entropy": 0.9592801034450531, |
| "epoch": 0.96, |
| "grad_norm": 0.11560477316379547, |
| "learning_rate": 0.0001, |
| "loss": 0.975141704082489, |
| "mean_token_accuracy": 0.7269536554813385, |
| "num_tokens": 3371272.0, |
| "step": 48 |
| }, |
| { |
| "entropy": 0.9194359481334686, |
| "epoch": 0.98, |
| "grad_norm": 0.11588755995035172, |
| "learning_rate": 0.0001, |
| "loss": 0.9375251531600952, |
| "mean_token_accuracy": 0.7386307418346405, |
| "num_tokens": 3439636.0, |
| "step": 49 |
| }, |
| { |
| "entropy": 0.9375229477882385, |
| "epoch": 1.0, |
| "grad_norm": 0.10675432533025742, |
| "learning_rate": 0.0001, |
| "loss": 0.9375709891319275, |
| "mean_token_accuracy": 0.7371015846729279, |
| "num_tokens": 3515643.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.9784637987613678, |
| "epoch": 1.02, |
| "grad_norm": 0.12157473713159561, |
| "learning_rate": 0.0001, |
| "loss": 0.9589431881904602, |
| "mean_token_accuracy": 0.7312025725841522, |
| "num_tokens": 3579936.0, |
| "step": 51 |
| }, |
| { |
| "entropy": 0.9976795613765717, |
| "epoch": 1.04, |
| "grad_norm": 0.1218739002943039, |
| "learning_rate": 0.0001, |
| "loss": 0.9651644825935364, |
| "mean_token_accuracy": 0.7295917272567749, |
| "num_tokens": 3645097.0, |
| "step": 52 |
| }, |
| { |
| "entropy": 0.9457567930221558, |
| "epoch": 1.06, |
| "grad_norm": 0.11789195984601974, |
| "learning_rate": 0.0001, |
| "loss": 0.924875020980835, |
| "mean_token_accuracy": 0.7387276589870453, |
| "num_tokens": 3722386.0, |
| "step": 53 |
| }, |
| { |
| "entropy": 0.9368592202663422, |
| "epoch": 1.08, |
| "grad_norm": 0.11367881298065186, |
| "learning_rate": 0.0001, |
| "loss": 0.9108861088752747, |
| "mean_token_accuracy": 0.7435753047466278, |
| "num_tokens": 3795107.0, |
| "step": 54 |
| }, |
| { |
| "entropy": 0.9856767952442169, |
| "epoch": 1.1, |
| "grad_norm": 0.1143391951918602, |
| "learning_rate": 0.0001, |
| "loss": 0.9522716999053955, |
| "mean_token_accuracy": 0.7351356148719788, |
| "num_tokens": 3863645.0, |
| "step": 55 |
| }, |
| { |
| "entropy": 0.9017468690872192, |
| "epoch": 1.12, |
| "grad_norm": 0.11213760077953339, |
| "learning_rate": 0.0001, |
| "loss": 0.890508770942688, |
| "mean_token_accuracy": 0.75059974193573, |
| "num_tokens": 3933157.0, |
| "step": 56 |
| }, |
| { |
| "entropy": 0.9076333045959473, |
| "epoch": 1.1400000000000001, |
| "grad_norm": 0.10830755531787872, |
| "learning_rate": 0.0001, |
| "loss": 0.8893829584121704, |
| "mean_token_accuracy": 0.7474276721477509, |
| "num_tokens": 4002549.0, |
| "step": 57 |
| }, |
| { |
| "entropy": 0.9090627431869507, |
| "epoch": 1.16, |
| "grad_norm": 0.11783402413129807, |
| "learning_rate": 0.0001, |
| "loss": 0.9097412824630737, |
| "mean_token_accuracy": 0.7423691749572754, |
| "num_tokens": 4068026.0, |
| "step": 58 |
| }, |
| { |
| "entropy": 0.928420215845108, |
| "epoch": 1.18, |
| "grad_norm": 0.11873085796833038, |
| "learning_rate": 0.0001, |
| "loss": 0.9404212236404419, |
| "mean_token_accuracy": 0.7362638711929321, |
| "num_tokens": 4136948.0, |
| "step": 59 |
| }, |
| { |
| "entropy": 0.9221147298812866, |
| "epoch": 1.2, |
| "grad_norm": 0.11550325155258179, |
| "learning_rate": 0.0001, |
| "loss": 0.924960732460022, |
| "mean_token_accuracy": 0.7427571713924408, |
| "num_tokens": 4204112.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.9216603636741638, |
| "epoch": 1.22, |
| "grad_norm": 0.11623509228229523, |
| "learning_rate": 0.0001, |
| "loss": 0.9106500148773193, |
| "mean_token_accuracy": 0.7429015040397644, |
| "num_tokens": 4272140.0, |
| "step": 61 |
| }, |
| { |
| "entropy": 0.9455150067806244, |
| "epoch": 1.24, |
| "grad_norm": 0.11591103672981262, |
| "learning_rate": 0.0001, |
| "loss": 0.9491208791732788, |
| "mean_token_accuracy": 0.7350127100944519, |
| "num_tokens": 4346769.0, |
| "step": 62 |
| }, |
| { |
| "entropy": 0.8898249566555023, |
| "epoch": 1.26, |
| "grad_norm": 0.11488021910190582, |
| "learning_rate": 0.0001, |
| "loss": 0.8970361351966858, |
| "mean_token_accuracy": 0.7481776773929596, |
| "num_tokens": 4418760.0, |
| "step": 63 |
| }, |
| { |
| "entropy": 0.8663885295391083, |
| "epoch": 1.28, |
| "grad_norm": 0.1245160847902298, |
| "learning_rate": 0.0001, |
| "loss": 0.8572047352790833, |
| "mean_token_accuracy": 0.7511683106422424, |
| "num_tokens": 4483453.0, |
| "step": 64 |
| }, |
| { |
| "entropy": 0.9115520715713501, |
| "epoch": 1.3, |
| "grad_norm": 0.12362005561590195, |
| "learning_rate": 0.0001, |
| "loss": 0.9021918773651123, |
| "mean_token_accuracy": 0.7433571219444275, |
| "num_tokens": 4553398.0, |
| "step": 65 |
| }, |
| { |
| "entropy": 0.9173152148723602, |
| "epoch": 1.32, |
| "grad_norm": 0.11844731867313385, |
| "learning_rate": 0.0001, |
| "loss": 0.9107658863067627, |
| "mean_token_accuracy": 0.7352526783943176, |
| "num_tokens": 4626227.0, |
| "step": 66 |
| }, |
| { |
| "entropy": 0.849992573261261, |
| "epoch": 1.34, |
| "grad_norm": 0.11784010380506516, |
| "learning_rate": 0.0001, |
| "loss": 0.8525710105895996, |
| "mean_token_accuracy": 0.7550307512283325, |
| "num_tokens": 4701977.0, |
| "step": 67 |
| }, |
| { |
| "entropy": 0.9553567171096802, |
| "epoch": 1.3599999999999999, |
| "grad_norm": 0.1190195307135582, |
| "learning_rate": 0.0001, |
| "loss": 0.9455575346946716, |
| "mean_token_accuracy": 0.7316735982894897, |
| "num_tokens": 4774545.0, |
| "step": 68 |
| }, |
| { |
| "entropy": 0.8593668639659882, |
| "epoch": 1.38, |
| "grad_norm": 0.11937589198350906, |
| "learning_rate": 0.0001, |
| "loss": 0.8549792170524597, |
| "mean_token_accuracy": 0.7569719552993774, |
| "num_tokens": 4847366.0, |
| "step": 69 |
| }, |
| { |
| "entropy": 0.9686211943626404, |
| "epoch": 1.4, |
| "grad_norm": 0.12214019149541855, |
| "learning_rate": 0.0001, |
| "loss": 0.9648027420043945, |
| "mean_token_accuracy": 0.7249570488929749, |
| "num_tokens": 4923060.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.867369145154953, |
| "epoch": 1.42, |
| "grad_norm": 0.11730711907148361, |
| "learning_rate": 0.0001, |
| "loss": 0.8510935306549072, |
| "mean_token_accuracy": 0.7553304731845856, |
| "num_tokens": 4996176.0, |
| "step": 71 |
| }, |
| { |
| "entropy": 0.9160025417804718, |
| "epoch": 1.44, |
| "grad_norm": 0.12190750241279602, |
| "learning_rate": 0.0001, |
| "loss": 0.9096644520759583, |
| "mean_token_accuracy": 0.7384364008903503, |
| "num_tokens": 5068522.0, |
| "step": 72 |
| }, |
| { |
| "entropy": 0.9069952070713043, |
| "epoch": 1.46, |
| "grad_norm": 0.13640820980072021, |
| "learning_rate": 0.0001, |
| "loss": 0.8875448703765869, |
| "mean_token_accuracy": 0.7457329034805298, |
| "num_tokens": 5137623.0, |
| "step": 73 |
| }, |
| { |
| "entropy": 0.9026708900928497, |
| "epoch": 1.48, |
| "grad_norm": 0.12538853287696838, |
| "learning_rate": 0.0001, |
| "loss": 0.8914889097213745, |
| "mean_token_accuracy": 0.7435721457004547, |
| "num_tokens": 5208737.0, |
| "step": 74 |
| }, |
| { |
| "entropy": 0.9324445724487305, |
| "epoch": 1.5, |
| "grad_norm": 0.13337835669517517, |
| "learning_rate": 0.0001, |
| "loss": 0.9412592649459839, |
| "mean_token_accuracy": 0.7320683300495148, |
| "num_tokens": 5279457.0, |
| "step": 75 |
| }, |
| { |
| "entropy": 0.8831305801868439, |
| "epoch": 1.52, |
| "grad_norm": 0.1287890523672104, |
| "learning_rate": 0.0001, |
| "loss": 0.890162467956543, |
| "mean_token_accuracy": 0.7474140524864197, |
| "num_tokens": 5344457.0, |
| "step": 76 |
| }, |
| { |
| "entropy": 0.8844259679317474, |
| "epoch": 1.54, |
| "grad_norm": 0.12595458328723907, |
| "learning_rate": 0.0001, |
| "loss": 0.8862929940223694, |
| "mean_token_accuracy": 0.7448306679725647, |
| "num_tokens": 5416486.0, |
| "step": 77 |
| }, |
| { |
| "entropy": 0.9768937230110168, |
| "epoch": 1.56, |
| "grad_norm": 0.13385823369026184, |
| "learning_rate": 0.0001, |
| "loss": 0.9855270981788635, |
| "mean_token_accuracy": 0.7242069244384766, |
| "num_tokens": 5488930.0, |
| "step": 78 |
| }, |
| { |
| "entropy": 0.8970398604869843, |
| "epoch": 1.58, |
| "grad_norm": 0.12776781618595123, |
| "learning_rate": 0.0001, |
| "loss": 0.9010695219039917, |
| "mean_token_accuracy": 0.7380562722682953, |
| "num_tokens": 5558044.0, |
| "step": 79 |
| }, |
| { |
| "entropy": 0.8510560393333435, |
| "epoch": 1.6, |
| "grad_norm": 0.12845851480960846, |
| "learning_rate": 0.0001, |
| "loss": 0.8674848675727844, |
| "mean_token_accuracy": 0.7530980706214905, |
| "num_tokens": 5630814.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.9194655120372772, |
| "epoch": 1.62, |
| "grad_norm": 0.12733094394207, |
| "learning_rate": 0.0001, |
| "loss": 0.9001493453979492, |
| "mean_token_accuracy": 0.7473941147327423, |
| "num_tokens": 5696496.0, |
| "step": 81 |
| }, |
| { |
| "entropy": 0.9078547358512878, |
| "epoch": 1.6400000000000001, |
| "grad_norm": 0.12357509136199951, |
| "learning_rate": 0.0001, |
| "loss": 0.9031519889831543, |
| "mean_token_accuracy": 0.7443413436412811, |
| "num_tokens": 5765578.0, |
| "step": 82 |
| }, |
| { |
| "entropy": 0.9426756799221039, |
| "epoch": 1.6600000000000001, |
| "grad_norm": 0.12899024784564972, |
| "learning_rate": 0.0001, |
| "loss": 0.9363412857055664, |
| "mean_token_accuracy": 0.7349725663661957, |
| "num_tokens": 5831979.0, |
| "step": 83 |
| }, |
| { |
| "entropy": 0.9146726131439209, |
| "epoch": 1.6800000000000002, |
| "grad_norm": 0.12931226193904877, |
| "learning_rate": 0.0001, |
| "loss": 0.9105992317199707, |
| "mean_token_accuracy": 0.7388072907924652, |
| "num_tokens": 5905661.0, |
| "step": 84 |
| }, |
| { |
| "entropy": 0.9316370487213135, |
| "epoch": 1.7, |
| "grad_norm": 0.13409629464149475, |
| "learning_rate": 0.0001, |
| "loss": 0.9174567461013794, |
| "mean_token_accuracy": 0.7437045872211456, |
| "num_tokens": 5973162.0, |
| "step": 85 |
| }, |
| { |
| "entropy": 0.9444582164287567, |
| "epoch": 1.72, |
| "grad_norm": 0.13663342595100403, |
| "learning_rate": 0.0001, |
| "loss": 0.9291839599609375, |
| "mean_token_accuracy": 0.7356755137443542, |
| "num_tokens": 6040751.0, |
| "step": 86 |
| }, |
| { |
| "entropy": 0.8949260711669922, |
| "epoch": 1.74, |
| "grad_norm": 0.12818805873394012, |
| "learning_rate": 0.0001, |
| "loss": 0.8870834112167358, |
| "mean_token_accuracy": 0.7440982460975647, |
| "num_tokens": 6112173.0, |
| "step": 87 |
| }, |
| { |
| "entropy": 0.9360968768596649, |
| "epoch": 1.76, |
| "grad_norm": 0.14662906527519226, |
| "learning_rate": 0.0001, |
| "loss": 0.9298614859580994, |
| "mean_token_accuracy": 0.7370602786540985, |
| "num_tokens": 6179601.0, |
| "step": 88 |
| }, |
| { |
| "entropy": 0.8710778653621674, |
| "epoch": 1.78, |
| "grad_norm": 0.13221921026706696, |
| "learning_rate": 0.0001, |
| "loss": 0.8649511337280273, |
| "mean_token_accuracy": 0.7522627413272858, |
| "num_tokens": 6246710.0, |
| "step": 89 |
| }, |
| { |
| "entropy": 0.8752692639827728, |
| "epoch": 1.8, |
| "grad_norm": 0.14055795967578888, |
| "learning_rate": 0.0001, |
| "loss": 0.8573565483093262, |
| "mean_token_accuracy": 0.754420816898346, |
| "num_tokens": 6319796.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.779484897851944, |
| "epoch": 1.8199999999999998, |
| "grad_norm": 0.12845420837402344, |
| "learning_rate": 0.0001, |
| "loss": 0.7835309505462646, |
| "mean_token_accuracy": 0.7740673124790192, |
| "num_tokens": 6389425.0, |
| "step": 91 |
| }, |
| { |
| "entropy": 0.8905294835567474, |
| "epoch": 1.8399999999999999, |
| "grad_norm": 0.1387389600276947, |
| "learning_rate": 0.0001, |
| "loss": 0.8962739706039429, |
| "mean_token_accuracy": 0.7440564632415771, |
| "num_tokens": 6460606.0, |
| "step": 92 |
| }, |
| { |
| "entropy": 0.8370858132839203, |
| "epoch": 1.8599999999999999, |
| "grad_norm": 0.13243098556995392, |
| "learning_rate": 0.0001, |
| "loss": 0.8395211696624756, |
| "mean_token_accuracy": 0.7600894570350647, |
| "num_tokens": 6532762.0, |
| "step": 93 |
| }, |
| { |
| "entropy": 0.9130788445472717, |
| "epoch": 1.88, |
| "grad_norm": 0.13279658555984497, |
| "learning_rate": 0.0001, |
| "loss": 0.9157933592796326, |
| "mean_token_accuracy": 0.7389859557151794, |
| "num_tokens": 6601952.0, |
| "step": 94 |
| }, |
| { |
| "entropy": 0.889918178319931, |
| "epoch": 1.9, |
| "grad_norm": 0.134121835231781, |
| "learning_rate": 0.0001, |
| "loss": 0.8863574266433716, |
| "mean_token_accuracy": 0.7509450018405914, |
| "num_tokens": 6667993.0, |
| "step": 95 |
| }, |
| { |
| "entropy": 0.8806619942188263, |
| "epoch": 1.92, |
| "grad_norm": 0.1342441886663437, |
| "learning_rate": 0.0001, |
| "loss": 0.8897389769554138, |
| "mean_token_accuracy": 0.751017302274704, |
| "num_tokens": 6741003.0, |
| "step": 96 |
| }, |
| { |
| "entropy": 0.8719884157180786, |
| "epoch": 1.94, |
| "grad_norm": 0.13280296325683594, |
| "learning_rate": 0.0001, |
| "loss": 0.8737912178039551, |
| "mean_token_accuracy": 0.748961865901947, |
| "num_tokens": 6814733.0, |
| "step": 97 |
| }, |
| { |
| "entropy": 0.9378086924552917, |
| "epoch": 1.96, |
| "grad_norm": 0.142800435423851, |
| "learning_rate": 0.0001, |
| "loss": 0.928570032119751, |
| "mean_token_accuracy": 0.7347837388515472, |
| "num_tokens": 6887723.0, |
| "step": 98 |
| }, |
| { |
| "entropy": 0.8699579238891602, |
| "epoch": 1.98, |
| "grad_norm": 0.13884496688842773, |
| "learning_rate": 0.0001, |
| "loss": 0.8652123808860779, |
| "mean_token_accuracy": 0.7550584077835083, |
| "num_tokens": 6957156.0, |
| "step": 99 |
| }, |
| { |
| "entropy": 0.8680048286914825, |
| "epoch": 2.0, |
| "grad_norm": 0.13625894486904144, |
| "learning_rate": 0.0001, |
| "loss": 0.8566628694534302, |
| "mean_token_accuracy": 0.7550990581512451, |
| "num_tokens": 7030824.0, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.8759163916110992, |
| "epoch": 2.02, |
| "grad_norm": 0.13642925024032593, |
| "learning_rate": 0.0001, |
| "loss": 0.8173877596855164, |
| "mean_token_accuracy": 0.7642171382904053, |
| "num_tokens": 7099794.0, |
| "step": 101 |
| }, |
| { |
| "entropy": 0.872030109167099, |
| "epoch": 2.04, |
| "grad_norm": 0.13444599509239197, |
| "learning_rate": 0.0001, |
| "loss": 0.8315644860267639, |
| "mean_token_accuracy": 0.7598301768302917, |
| "num_tokens": 7168851.0, |
| "step": 102 |
| }, |
| { |
| "entropy": 0.8999165594577789, |
| "epoch": 2.06, |
| "grad_norm": 0.14029845595359802, |
| "learning_rate": 0.0001, |
| "loss": 0.8543646335601807, |
| "mean_token_accuracy": 0.7513293921947479, |
| "num_tokens": 7239717.0, |
| "step": 103 |
| }, |
| { |
| "entropy": 0.8488370180130005, |
| "epoch": 2.08, |
| "grad_norm": 0.13906973600387573, |
| "learning_rate": 0.0001, |
| "loss": 0.8071736097335815, |
| "mean_token_accuracy": 0.768627405166626, |
| "num_tokens": 7310897.0, |
| "step": 104 |
| }, |
| { |
| "entropy": 0.8458687663078308, |
| "epoch": 2.1, |
| "grad_norm": 0.14410872757434845, |
| "learning_rate": 0.0001, |
| "loss": 0.8336784243583679, |
| "mean_token_accuracy": 0.7576306164264679, |
| "num_tokens": 7381044.0, |
| "step": 105 |
| }, |
| { |
| "entropy": 0.8117028772830963, |
| "epoch": 2.12, |
| "grad_norm": 0.15016138553619385, |
| "learning_rate": 0.0001, |
| "loss": 0.8016491532325745, |
| "mean_token_accuracy": 0.761463850736618, |
| "num_tokens": 7448539.0, |
| "step": 106 |
| }, |
| { |
| "entropy": 0.8423402607440948, |
| "epoch": 2.14, |
| "grad_norm": 0.14620108902454376, |
| "learning_rate": 0.0001, |
| "loss": 0.8520156145095825, |
| "mean_token_accuracy": 0.7559936344623566, |
| "num_tokens": 7523326.0, |
| "step": 107 |
| }, |
| { |
| "entropy": 0.7789672613143921, |
| "epoch": 2.16, |
| "grad_norm": 0.14582377672195435, |
| "learning_rate": 0.0001, |
| "loss": 0.7745522260665894, |
| "mean_token_accuracy": 0.7746096849441528, |
| "num_tokens": 7589964.0, |
| "step": 108 |
| }, |
| { |
| "entropy": 0.8252202570438385, |
| "epoch": 2.18, |
| "grad_norm": 0.15269358456134796, |
| "learning_rate": 0.0001, |
| "loss": 0.8253834247589111, |
| "mean_token_accuracy": 0.760587066411972, |
| "num_tokens": 7663822.0, |
| "step": 109 |
| }, |
| { |
| "entropy": 0.8460159599781036, |
| "epoch": 2.2, |
| "grad_norm": 0.15312956273555756, |
| "learning_rate": 0.0001, |
| "loss": 0.834621012210846, |
| "mean_token_accuracy": 0.7548129558563232, |
| "num_tokens": 7736094.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.8546044528484344, |
| "epoch": 2.22, |
| "grad_norm": 0.1496986299753189, |
| "learning_rate": 0.0001, |
| "loss": 0.8300120830535889, |
| "mean_token_accuracy": 0.759436160326004, |
| "num_tokens": 7812027.0, |
| "step": 111 |
| }, |
| { |
| "entropy": 0.754067599773407, |
| "epoch": 2.24, |
| "grad_norm": 0.1404401957988739, |
| "learning_rate": 0.0001, |
| "loss": 0.7299779653549194, |
| "mean_token_accuracy": 0.7858331501483917, |
| "num_tokens": 7881889.0, |
| "step": 112 |
| }, |
| { |
| "entropy": 0.8271636366844177, |
| "epoch": 2.26, |
| "grad_norm": 0.15130330622196198, |
| "learning_rate": 0.0001, |
| "loss": 0.8026934862136841, |
| "mean_token_accuracy": 0.7606152594089508, |
| "num_tokens": 7958772.0, |
| "step": 113 |
| }, |
| { |
| "entropy": 0.8311681747436523, |
| "epoch": 2.2800000000000002, |
| "grad_norm": 0.15545408427715302, |
| "learning_rate": 0.0001, |
| "loss": 0.8179649114608765, |
| "mean_token_accuracy": 0.7627213597297668, |
| "num_tokens": 8029445.0, |
| "step": 114 |
| }, |
| { |
| "entropy": 0.8371923565864563, |
| "epoch": 2.3, |
| "grad_norm": 0.15620867908000946, |
| "learning_rate": 0.0001, |
| "loss": 0.8322392106056213, |
| "mean_token_accuracy": 0.7557950615882874, |
| "num_tokens": 8100496.0, |
| "step": 115 |
| }, |
| { |
| "entropy": 0.8774853646755219, |
| "epoch": 2.32, |
| "grad_norm": 0.16020546853542328, |
| "learning_rate": 0.0001, |
| "loss": 0.8756296634674072, |
| "mean_token_accuracy": 0.7484939396381378, |
| "num_tokens": 8172797.0, |
| "step": 116 |
| }, |
| { |
| "entropy": 0.7841235101222992, |
| "epoch": 2.34, |
| "grad_norm": 0.16341933608055115, |
| "learning_rate": 0.0001, |
| "loss": 0.7845247387886047, |
| "mean_token_accuracy": 0.7680597603321075, |
| "num_tokens": 8239973.0, |
| "step": 117 |
| }, |
| { |
| "entropy": 0.8059554100036621, |
| "epoch": 2.36, |
| "grad_norm": 0.14973363280296326, |
| "learning_rate": 0.0001, |
| "loss": 0.7868502140045166, |
| "mean_token_accuracy": 0.7696151733398438, |
| "num_tokens": 8312997.0, |
| "step": 118 |
| }, |
| { |
| "entropy": 0.790012925863266, |
| "epoch": 2.38, |
| "grad_norm": 0.15735657513141632, |
| "learning_rate": 0.0001, |
| "loss": 0.7770582437515259, |
| "mean_token_accuracy": 0.7715879082679749, |
| "num_tokens": 8380963.0, |
| "step": 119 |
| }, |
| { |
| "entropy": 0.7878563702106476, |
| "epoch": 2.4, |
| "grad_norm": 0.15908536314964294, |
| "learning_rate": 0.0001, |
| "loss": 0.7946950197219849, |
| "mean_token_accuracy": 0.7720945775508881, |
| "num_tokens": 8453009.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.8163405060768127, |
| "epoch": 2.42, |
| "grad_norm": 0.15914228558540344, |
| "learning_rate": 0.0001, |
| "loss": 0.8022626042366028, |
| "mean_token_accuracy": 0.765992671251297, |
| "num_tokens": 8522525.0, |
| "step": 121 |
| }, |
| { |
| "entropy": 0.8679061532020569, |
| "epoch": 2.44, |
| "grad_norm": 0.1549653708934784, |
| "learning_rate": 0.0001, |
| "loss": 0.8452656269073486, |
| "mean_token_accuracy": 0.7564500868320465, |
| "num_tokens": 8591511.0, |
| "step": 122 |
| }, |
| { |
| "entropy": 0.8327716588973999, |
| "epoch": 2.46, |
| "grad_norm": 0.16247446835041046, |
| "learning_rate": 0.0001, |
| "loss": 0.8243684768676758, |
| "mean_token_accuracy": 0.7590730786323547, |
| "num_tokens": 8665286.0, |
| "step": 123 |
| }, |
| { |
| "entropy": 0.7765352427959442, |
| "epoch": 2.48, |
| "grad_norm": 0.15758228302001953, |
| "learning_rate": 0.0001, |
| "loss": 0.7422535419464111, |
| "mean_token_accuracy": 0.783282607793808, |
| "num_tokens": 8730146.0, |
| "step": 124 |
| }, |
| { |
| "entropy": 0.8039740025997162, |
| "epoch": 2.5, |
| "grad_norm": 0.1573210507631302, |
| "learning_rate": 0.0001, |
| "loss": 0.7967308759689331, |
| "mean_token_accuracy": 0.7692459225654602, |
| "num_tokens": 8801629.0, |
| "step": 125 |
| }, |
| { |
| "entropy": 0.8415170609951019, |
| "epoch": 2.52, |
| "grad_norm": 0.16268476843833923, |
| "learning_rate": 0.0001, |
| "loss": 0.8165377378463745, |
| "mean_token_accuracy": 0.7615750133991241, |
| "num_tokens": 8877985.0, |
| "step": 126 |
| }, |
| { |
| "entropy": 0.8203210234642029, |
| "epoch": 2.54, |
| "grad_norm": 0.16346436738967896, |
| "learning_rate": 0.0001, |
| "loss": 0.8121747970581055, |
| "mean_token_accuracy": 0.761315256357193, |
| "num_tokens": 8951559.0, |
| "step": 127 |
| }, |
| { |
| "entropy": 0.8293347358703613, |
| "epoch": 2.56, |
| "grad_norm": 0.17023281753063202, |
| "learning_rate": 0.0001, |
| "loss": 0.8173530101776123, |
| "mean_token_accuracy": 0.759048581123352, |
| "num_tokens": 9021509.0, |
| "step": 128 |
| }, |
| { |
| "entropy": 0.7746993601322174, |
| "epoch": 2.58, |
| "grad_norm": 0.1615564227104187, |
| "learning_rate": 0.0001, |
| "loss": 0.7658690214157104, |
| "mean_token_accuracy": 0.7726005017757416, |
| "num_tokens": 9090508.0, |
| "step": 129 |
| }, |
| { |
| "entropy": 0.8143576085567474, |
| "epoch": 2.6, |
| "grad_norm": 0.1704084575176239, |
| "learning_rate": 0.0001, |
| "loss": 0.8276199102401733, |
| "mean_token_accuracy": 0.7586769461631775, |
| "num_tokens": 9156045.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.7868927419185638, |
| "epoch": 2.62, |
| "grad_norm": 0.16065886616706848, |
| "learning_rate": 0.0001, |
| "loss": 0.7649659514427185, |
| "mean_token_accuracy": 0.7774228751659393, |
| "num_tokens": 9221515.0, |
| "step": 131 |
| }, |
| { |
| "entropy": 0.8241930305957794, |
| "epoch": 2.64, |
| "grad_norm": 0.1710738241672516, |
| "learning_rate": 0.0001, |
| "loss": 0.8104192018508911, |
| "mean_token_accuracy": 0.7651428282260895, |
| "num_tokens": 9290829.0, |
| "step": 132 |
| }, |
| { |
| "entropy": 0.7782109975814819, |
| "epoch": 2.66, |
| "grad_norm": 0.3252999484539032, |
| "learning_rate": 0.0001, |
| "loss": 0.7537873983383179, |
| "mean_token_accuracy": 0.7770578563213348, |
| "num_tokens": 9358923.0, |
| "step": 133 |
| }, |
| { |
| "entropy": 0.7932650744915009, |
| "epoch": 2.68, |
| "grad_norm": 0.16687457263469696, |
| "learning_rate": 0.0001, |
| "loss": 0.771995484828949, |
| "mean_token_accuracy": 0.7717720866203308, |
| "num_tokens": 9431495.0, |
| "step": 134 |
| }, |
| { |
| "entropy": 0.8635589182376862, |
| "epoch": 2.7, |
| "grad_norm": 0.17788194119930267, |
| "learning_rate": 0.0001, |
| "loss": 0.839271605014801, |
| "mean_token_accuracy": 0.7549299001693726, |
| "num_tokens": 9502949.0, |
| "step": 135 |
| }, |
| { |
| "entropy": 0.8827160894870758, |
| "epoch": 2.7199999999999998, |
| "grad_norm": 0.17214906215667725, |
| "learning_rate": 0.0001, |
| "loss": 0.8584389686584473, |
| "mean_token_accuracy": 0.7523661851882935, |
| "num_tokens": 9572738.0, |
| "step": 136 |
| }, |
| { |
| "entropy": 0.7438818514347076, |
| "epoch": 2.74, |
| "grad_norm": 0.15866513550281525, |
| "learning_rate": 0.0001, |
| "loss": 0.7251607179641724, |
| "mean_token_accuracy": 0.7865616679191589, |
| "num_tokens": 9643339.0, |
| "step": 137 |
| }, |
| { |
| "entropy": 0.8008526265621185, |
| "epoch": 2.76, |
| "grad_norm": 0.17728553712368011, |
| "learning_rate": 0.0001, |
| "loss": 0.7907190322875977, |
| "mean_token_accuracy": 0.7696272730827332, |
| "num_tokens": 9712087.0, |
| "step": 138 |
| }, |
| { |
| "entropy": 0.7775652408599854, |
| "epoch": 2.7800000000000002, |
| "grad_norm": 0.16898134350776672, |
| "learning_rate": 0.0001, |
| "loss": 0.7762832641601562, |
| "mean_token_accuracy": 0.7671337127685547, |
| "num_tokens": 9787274.0, |
| "step": 139 |
| }, |
| { |
| "entropy": 0.7524736523628235, |
| "epoch": 2.8, |
| "grad_norm": 0.1687132865190506, |
| "learning_rate": 0.0001, |
| "loss": 0.7515966892242432, |
| "mean_token_accuracy": 0.7743718028068542, |
| "num_tokens": 9850649.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.784816324710846, |
| "epoch": 2.82, |
| "grad_norm": 0.16266104578971863, |
| "learning_rate": 0.0001, |
| "loss": 0.7850303649902344, |
| "mean_token_accuracy": 0.7671224176883698, |
| "num_tokens": 9919647.0, |
| "step": 141 |
| }, |
| { |
| "entropy": 0.845661848783493, |
| "epoch": 2.84, |
| "grad_norm": 0.177549347281456, |
| "learning_rate": 0.0001, |
| "loss": 0.8321537375450134, |
| "mean_token_accuracy": 0.7545971870422363, |
| "num_tokens": 9988951.0, |
| "step": 142 |
| }, |
| { |
| "entropy": 0.7890813946723938, |
| "epoch": 2.86, |
| "grad_norm": 0.16998162865638733, |
| "learning_rate": 0.0001, |
| "loss": 0.7851375937461853, |
| "mean_token_accuracy": 0.77273029088974, |
| "num_tokens": 10059418.0, |
| "step": 143 |
| }, |
| { |
| "entropy": 0.8383583724498749, |
| "epoch": 2.88, |
| "grad_norm": 0.16811728477478027, |
| "learning_rate": 0.0001, |
| "loss": 0.8264563083648682, |
| "mean_token_accuracy": 0.7561212778091431, |
| "num_tokens": 10130582.0, |
| "step": 144 |
| }, |
| { |
| "entropy": 0.7998828291893005, |
| "epoch": 2.9, |
| "grad_norm": 0.162356436252594, |
| "learning_rate": 0.0001, |
| "loss": 0.7867046594619751, |
| "mean_token_accuracy": 0.7678299248218536, |
| "num_tokens": 10205184.0, |
| "step": 145 |
| }, |
| { |
| "entropy": 0.8201400637626648, |
| "epoch": 2.92, |
| "grad_norm": 0.1775413304567337, |
| "learning_rate": 0.0001, |
| "loss": 0.8155138492584229, |
| "mean_token_accuracy": 0.7631427645683289, |
| "num_tokens": 10272128.0, |
| "step": 146 |
| }, |
| { |
| "entropy": 0.8001563251018524, |
| "epoch": 2.94, |
| "grad_norm": 0.1714172512292862, |
| "learning_rate": 0.0001, |
| "loss": 0.7718967795372009, |
| "mean_token_accuracy": 0.7731671333312988, |
| "num_tokens": 10339140.0, |
| "step": 147 |
| }, |
| { |
| "entropy": 0.8686825633049011, |
| "epoch": 2.96, |
| "grad_norm": 0.1718156337738037, |
| "learning_rate": 0.0001, |
| "loss": 0.841813862323761, |
| "mean_token_accuracy": 0.7538741528987885, |
| "num_tokens": 10405402.0, |
| "step": 148 |
| }, |
| { |
| "entropy": 0.8255911767482758, |
| "epoch": 2.98, |
| "grad_norm": 0.17787101864814758, |
| "learning_rate": 0.0001, |
| "loss": 0.7991260290145874, |
| "mean_token_accuracy": 0.7636489570140839, |
| "num_tokens": 10473811.0, |
| "step": 149 |
| }, |
| { |
| "entropy": 0.8001417219638824, |
| "epoch": 3.0, |
| "grad_norm": 0.17027507722377777, |
| "learning_rate": 0.0001, |
| "loss": 0.7719438076019287, |
| "mean_token_accuracy": 0.7698873281478882, |
| "num_tokens": 10546089.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.8200977146625519, |
| "epoch": 3.02, |
| "grad_norm": 0.16800017654895782, |
| "learning_rate": 0.0001, |
| "loss": 0.7360785007476807, |
| "mean_token_accuracy": 0.7826903462409973, |
| "num_tokens": 10618015.0, |
| "step": 151 |
| }, |
| { |
| "entropy": 0.8175333142280579, |
| "epoch": 3.04, |
| "grad_norm": 0.18896523118019104, |
| "learning_rate": 0.0001, |
| "loss": 0.7504415512084961, |
| "mean_token_accuracy": 0.7746379971504211, |
| "num_tokens": 10688911.0, |
| "step": 152 |
| }, |
| { |
| "entropy": 0.7592836618423462, |
| "epoch": 3.06, |
| "grad_norm": 0.18069066107273102, |
| "learning_rate": 0.0001, |
| "loss": 0.7176339626312256, |
| "mean_token_accuracy": 0.7862937152385712, |
| "num_tokens": 10755190.0, |
| "step": 153 |
| }, |
| { |
| "entropy": 0.7064773738384247, |
| "epoch": 3.08, |
| "grad_norm": 0.17506442964076996, |
| "learning_rate": 0.0001, |
| "loss": 0.6661837697029114, |
| "mean_token_accuracy": 0.7990798652172089, |
| "num_tokens": 10824572.0, |
| "step": 154 |
| }, |
| { |
| "entropy": 0.7049046158790588, |
| "epoch": 3.1, |
| "grad_norm": 0.18128055334091187, |
| "learning_rate": 0.0001, |
| "loss": 0.6970986127853394, |
| "mean_token_accuracy": 0.7910968661308289, |
| "num_tokens": 10897921.0, |
| "step": 155 |
| }, |
| { |
| "entropy": 0.7131330072879791, |
| "epoch": 3.12, |
| "grad_norm": 0.1793287843465805, |
| "learning_rate": 0.0001, |
| "loss": 0.6969539523124695, |
| "mean_token_accuracy": 0.7904528379440308, |
| "num_tokens": 10967473.0, |
| "step": 156 |
| }, |
| { |
| "entropy": 0.7718028426170349, |
| "epoch": 3.14, |
| "grad_norm": 0.18882189691066742, |
| "learning_rate": 0.0001, |
| "loss": 0.7447178959846497, |
| "mean_token_accuracy": 0.7770456969738007, |
| "num_tokens": 11040256.0, |
| "step": 157 |
| }, |
| { |
| "entropy": 0.753845751285553, |
| "epoch": 3.16, |
| "grad_norm": 0.20335878431797028, |
| "learning_rate": 0.0001, |
| "loss": 0.7178476452827454, |
| "mean_token_accuracy": 0.7853229939937592, |
| "num_tokens": 11106853.0, |
| "step": 158 |
| }, |
| { |
| "entropy": 0.6985587477684021, |
| "epoch": 3.18, |
| "grad_norm": 0.18876291811466217, |
| "learning_rate": 0.0001, |
| "loss": 0.6741787195205688, |
| "mean_token_accuracy": 0.7952553629875183, |
| "num_tokens": 11178868.0, |
| "step": 159 |
| }, |
| { |
| "entropy": 0.7184737622737885, |
| "epoch": 3.2, |
| "grad_norm": 0.19795499742031097, |
| "learning_rate": 0.0001, |
| "loss": 0.6854847073554993, |
| "mean_token_accuracy": 0.7917522490024567, |
| "num_tokens": 11246484.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.7506313323974609, |
| "epoch": 3.22, |
| "grad_norm": 0.2013498693704605, |
| "learning_rate": 0.0001, |
| "loss": 0.7236944437026978, |
| "mean_token_accuracy": 0.7839900851249695, |
| "num_tokens": 11319820.0, |
| "step": 161 |
| }, |
| { |
| "entropy": 0.7317833006381989, |
| "epoch": 3.24, |
| "grad_norm": 0.19697904586791992, |
| "learning_rate": 0.0001, |
| "loss": 0.7071319222450256, |
| "mean_token_accuracy": 0.7910904288291931, |
| "num_tokens": 11389201.0, |
| "step": 162 |
| }, |
| { |
| "entropy": 0.7270588874816895, |
| "epoch": 3.26, |
| "grad_norm": 0.19727207720279694, |
| "learning_rate": 0.0001, |
| "loss": 0.7034903168678284, |
| "mean_token_accuracy": 0.7924522161483765, |
| "num_tokens": 11458094.0, |
| "step": 163 |
| }, |
| { |
| "entropy": 0.6620264649391174, |
| "epoch": 3.2800000000000002, |
| "grad_norm": 0.1949145644903183, |
| "learning_rate": 0.0001, |
| "loss": 0.6431432366371155, |
| "mean_token_accuracy": 0.8045243918895721, |
| "num_tokens": 11526439.0, |
| "step": 164 |
| }, |
| { |
| "entropy": 0.7114729583263397, |
| "epoch": 3.3, |
| "grad_norm": 0.2044411301612854, |
| "learning_rate": 0.0001, |
| "loss": 0.6916588544845581, |
| "mean_token_accuracy": 0.7911301851272583, |
| "num_tokens": 11593907.0, |
| "step": 165 |
| }, |
| { |
| "entropy": 0.7394631803035736, |
| "epoch": 3.32, |
| "grad_norm": 0.20598116517066956, |
| "learning_rate": 0.0001, |
| "loss": 0.7143540978431702, |
| "mean_token_accuracy": 0.7861869037151337, |
| "num_tokens": 11659313.0, |
| "step": 166 |
| }, |
| { |
| "entropy": 0.7148211598396301, |
| "epoch": 3.34, |
| "grad_norm": 0.19798815250396729, |
| "learning_rate": 0.0001, |
| "loss": 0.6806311011314392, |
| "mean_token_accuracy": 0.7966620624065399, |
| "num_tokens": 11729203.0, |
| "step": 167 |
| }, |
| { |
| "entropy": 0.7278057634830475, |
| "epoch": 3.36, |
| "grad_norm": 0.2106526643037796, |
| "learning_rate": 0.0001, |
| "loss": 0.6907294392585754, |
| "mean_token_accuracy": 0.7917556762695312, |
| "num_tokens": 11798662.0, |
| "step": 168 |
| }, |
| { |
| "entropy": 0.7575972974300385, |
| "epoch": 3.38, |
| "grad_norm": 0.19657012820243835, |
| "learning_rate": 0.0001, |
| "loss": 0.7215894460678101, |
| "mean_token_accuracy": 0.7850257754325867, |
| "num_tokens": 11874700.0, |
| "step": 169 |
| }, |
| { |
| "entropy": 0.7198713719844818, |
| "epoch": 3.4, |
| "grad_norm": 0.19392815232276917, |
| "learning_rate": 0.0001, |
| "loss": 0.6976982355117798, |
| "mean_token_accuracy": 0.7894782423973083, |
| "num_tokens": 11946770.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.6691116690635681, |
| "epoch": 3.42, |
| "grad_norm": 0.1919872909784317, |
| "learning_rate": 0.0001, |
| "loss": 0.6429901123046875, |
| "mean_token_accuracy": 0.8067097663879395, |
| "num_tokens": 12016943.0, |
| "step": 171 |
| }, |
| { |
| "entropy": 0.7087913751602173, |
| "epoch": 3.44, |
| "grad_norm": 0.19245345890522003, |
| "learning_rate": 0.0001, |
| "loss": 0.6743010878562927, |
| "mean_token_accuracy": 0.8007382750511169, |
| "num_tokens": 12088223.0, |
| "step": 172 |
| }, |
| { |
| "entropy": 0.7334369122982025, |
| "epoch": 3.46, |
| "grad_norm": 0.1944996863603592, |
| "learning_rate": 0.0001, |
| "loss": 0.7241754531860352, |
| "mean_token_accuracy": 0.781945675611496, |
| "num_tokens": 12164031.0, |
| "step": 173 |
| }, |
| { |
| "entropy": 0.7216465771198273, |
| "epoch": 3.48, |
| "grad_norm": 0.195100337266922, |
| "learning_rate": 0.0001, |
| "loss": 0.6905370950698853, |
| "mean_token_accuracy": 0.7926844656467438, |
| "num_tokens": 12235488.0, |
| "step": 174 |
| }, |
| { |
| "entropy": 0.7634322941303253, |
| "epoch": 3.5, |
| "grad_norm": 0.20126116275787354, |
| "learning_rate": 0.0001, |
| "loss": 0.7199342250823975, |
| "mean_token_accuracy": 0.784062385559082, |
| "num_tokens": 12304402.0, |
| "step": 175 |
| }, |
| { |
| "entropy": 0.7162297964096069, |
| "epoch": 3.52, |
| "grad_norm": 0.19708800315856934, |
| "learning_rate": 0.0001, |
| "loss": 0.6828768253326416, |
| "mean_token_accuracy": 0.7963770925998688, |
| "num_tokens": 12376874.0, |
| "step": 176 |
| }, |
| { |
| "entropy": 0.7292175590991974, |
| "epoch": 3.54, |
| "grad_norm": 0.19853827357292175, |
| "learning_rate": 0.0001, |
| "loss": 0.6944613456726074, |
| "mean_token_accuracy": 0.7901931405067444, |
| "num_tokens": 12443682.0, |
| "step": 177 |
| }, |
| { |
| "entropy": 0.7690876424312592, |
| "epoch": 3.56, |
| "grad_norm": 0.2102740854024887, |
| "learning_rate": 0.0001, |
| "loss": 0.7348027229309082, |
| "mean_token_accuracy": 0.7807092070579529, |
| "num_tokens": 12511430.0, |
| "step": 178 |
| }, |
| { |
| "entropy": 0.7362220287322998, |
| "epoch": 3.58, |
| "grad_norm": 0.20193730294704437, |
| "learning_rate": 0.0001, |
| "loss": 0.7197112441062927, |
| "mean_token_accuracy": 0.7791264057159424, |
| "num_tokens": 12582173.0, |
| "step": 179 |
| }, |
| { |
| "entropy": 0.716655433177948, |
| "epoch": 3.6, |
| "grad_norm": 0.19548293948173523, |
| "learning_rate": 0.0001, |
| "loss": 0.7020804286003113, |
| "mean_token_accuracy": 0.789775550365448, |
| "num_tokens": 12655124.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.7400652468204498, |
| "epoch": 3.62, |
| "grad_norm": 0.20551350712776184, |
| "learning_rate": 0.0001, |
| "loss": 0.7202802300453186, |
| "mean_token_accuracy": 0.7863712012767792, |
| "num_tokens": 12728612.0, |
| "step": 181 |
| }, |
| { |
| "entropy": 0.7541409730911255, |
| "epoch": 3.64, |
| "grad_norm": 0.200267493724823, |
| "learning_rate": 0.0001, |
| "loss": 0.713320255279541, |
| "mean_token_accuracy": 0.7859389185905457, |
| "num_tokens": 12800056.0, |
| "step": 182 |
| }, |
| { |
| "entropy": 0.6940381526947021, |
| "epoch": 3.66, |
| "grad_norm": 0.19558116793632507, |
| "learning_rate": 0.0001, |
| "loss": 0.6574203968048096, |
| "mean_token_accuracy": 0.8014397621154785, |
| "num_tokens": 12868372.0, |
| "step": 183 |
| }, |
| { |
| "entropy": 0.7630845606327057, |
| "epoch": 3.68, |
| "grad_norm": 0.20073311030864716, |
| "learning_rate": 0.0001, |
| "loss": 0.7155758738517761, |
| "mean_token_accuracy": 0.7896965146064758, |
| "num_tokens": 12937518.0, |
| "step": 184 |
| }, |
| { |
| "entropy": 0.7708411812782288, |
| "epoch": 3.7, |
| "grad_norm": 0.22150248289108276, |
| "learning_rate": 0.0001, |
| "loss": 0.7369881868362427, |
| "mean_token_accuracy": 0.7802594602108002, |
| "num_tokens": 13003915.0, |
| "step": 185 |
| }, |
| { |
| "entropy": 0.7173754870891571, |
| "epoch": 3.7199999999999998, |
| "grad_norm": 0.19912731647491455, |
| "learning_rate": 0.0001, |
| "loss": 0.6876958608627319, |
| "mean_token_accuracy": 0.7933018803596497, |
| "num_tokens": 13075561.0, |
| "step": 186 |
| }, |
| { |
| "entropy": 0.7382989227771759, |
| "epoch": 3.74, |
| "grad_norm": 0.20131585001945496, |
| "learning_rate": 0.0001, |
| "loss": 0.7099729180335999, |
| "mean_token_accuracy": 0.7907689809799194, |
| "num_tokens": 13147726.0, |
| "step": 187 |
| }, |
| { |
| "entropy": 0.7296792566776276, |
| "epoch": 3.76, |
| "grad_norm": 0.19849984347820282, |
| "learning_rate": 0.0001, |
| "loss": 0.7037006616592407, |
| "mean_token_accuracy": 0.7868267595767975, |
| "num_tokens": 13219764.0, |
| "step": 188 |
| }, |
| { |
| "entropy": 0.7335399687290192, |
| "epoch": 3.7800000000000002, |
| "grad_norm": 0.20115980505943298, |
| "learning_rate": 0.0001, |
| "loss": 0.7100788354873657, |
| "mean_token_accuracy": 0.786194235086441, |
| "num_tokens": 13291698.0, |
| "step": 189 |
| }, |
| { |
| "entropy": 0.7148082256317139, |
| "epoch": 3.8, |
| "grad_norm": 0.19843769073486328, |
| "learning_rate": 0.0001, |
| "loss": 0.6846855878829956, |
| "mean_token_accuracy": 0.7930681705474854, |
| "num_tokens": 13364738.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.71998131275177, |
| "epoch": 3.82, |
| "grad_norm": 0.20854182541370392, |
| "learning_rate": 0.0001, |
| "loss": 0.6897561550140381, |
| "mean_token_accuracy": 0.7933003902435303, |
| "num_tokens": 13436984.0, |
| "step": 191 |
| }, |
| { |
| "entropy": 0.7520318031311035, |
| "epoch": 3.84, |
| "grad_norm": 0.21299579739570618, |
| "learning_rate": 0.0001, |
| "loss": 0.7223616242408752, |
| "mean_token_accuracy": 0.7838756442070007, |
| "num_tokens": 13504383.0, |
| "step": 192 |
| }, |
| { |
| "entropy": 0.7173813283443451, |
| "epoch": 3.86, |
| "grad_norm": 0.20238681137561798, |
| "learning_rate": 0.0001, |
| "loss": 0.6963660717010498, |
| "mean_token_accuracy": 0.7877775430679321, |
| "num_tokens": 13578910.0, |
| "step": 193 |
| }, |
| { |
| "entropy": 0.7405160069465637, |
| "epoch": 3.88, |
| "grad_norm": 0.21617071330547333, |
| "learning_rate": 0.0001, |
| "loss": 0.7227989435195923, |
| "mean_token_accuracy": 0.782155841588974, |
| "num_tokens": 13646579.0, |
| "step": 194 |
| }, |
| { |
| "entropy": 0.6996143460273743, |
| "epoch": 3.9, |
| "grad_norm": 0.20972195267677307, |
| "learning_rate": 0.0001, |
| "loss": 0.6725365519523621, |
| "mean_token_accuracy": 0.7987044751644135, |
| "num_tokens": 13712265.0, |
| "step": 195 |
| }, |
| { |
| "entropy": 0.6924785077571869, |
| "epoch": 3.92, |
| "grad_norm": 0.20143532752990723, |
| "learning_rate": 0.0001, |
| "loss": 0.6744241118431091, |
| "mean_token_accuracy": 0.7988313138484955, |
| "num_tokens": 13778938.0, |
| "step": 196 |
| }, |
| { |
| "entropy": 0.7586347460746765, |
| "epoch": 3.94, |
| "grad_norm": 0.21675406396389008, |
| "learning_rate": 0.0001, |
| "loss": 0.720883846282959, |
| "mean_token_accuracy": 0.7846451103687286, |
| "num_tokens": 13850680.0, |
| "step": 197 |
| }, |
| { |
| "entropy": 0.7003247439861298, |
| "epoch": 3.96, |
| "grad_norm": 0.20092730224132538, |
| "learning_rate": 0.0001, |
| "loss": 0.6743336915969849, |
| "mean_token_accuracy": 0.797335296869278, |
| "num_tokens": 13922486.0, |
| "step": 198 |
| }, |
| { |
| "entropy": 0.7424158155918121, |
| "epoch": 3.98, |
| "grad_norm": 0.2068347930908203, |
| "learning_rate": 0.0001, |
| "loss": 0.7222142815589905, |
| "mean_token_accuracy": 0.7855294048786163, |
| "num_tokens": 13994794.0, |
| "step": 199 |
| }, |
| { |
| "entropy": 0.7328775823116302, |
| "epoch": 4.0, |
| "grad_norm": 0.21473339200019836, |
| "learning_rate": 0.0001, |
| "loss": 0.6824338436126709, |
| "mean_token_accuracy": 0.7947524189949036, |
| "num_tokens": 14060764.0, |
| "step": 200 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 200, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 4, |
| "save_steps": 50, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 8.561776579712123e+17, |
| "train_batch_size": 8, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
